No abstract

No abstract

Axial flow blood pumps for cardiac assistance have proven their clinical viability and benefit in recent years. However, the clinical systems to date have no direct mechanism to decrease pump speed when adequate supply is not available. This may lead to ventricular collapse or increase the probability of hemolysis and thrombotic risks. Based on various experiences with left ventricular assist device (LVAD) patients in various states of recovery, at implant, in the intensive care unit, in the standard ward, and during physical exercise, 11 different algorithms were developed for the automatic detection of ventricular suction. These detection algorithms analyze the flow pattern for the presence of distinct suction indicators. For selection and optimization of the algorithms, 1000 records from approximately 100 patients were collected. Each record contains 5 s of pump flow, current, and arterial pressure. Three experts classified these records in terms of suction probability and other abnormalities. The optimization was developed in Matlab, capable of solving a fifth-dimensional optimization problem with 256 different algorithm combinations. The optimization resulted in a set of 6 algorithms, each with specific thresholds. The system detects 100% of the known suction events with 0.28% of false-positive interpretations. If tuned to avoid any false-positive detection, 90.7% of the certain events would be detected. A strategy for the development of a robust suction detection system for axial blood pumps was found. This system will be integrated into an automatic pump speed control system to provide adequate perfusion for the LVAD recipient, without excessive unloading of the ventricle.

The MicroMed DeBakey Ventricular Assist Device (MicroMed Technology, Inc., Houston, TX) is a continuous axial flow pump designed for long-term circulatory support. The system received CE approval in 2001 as a bridge to transplantation and in 2004 as an alternative to transplantation. Low volume in the left ventricle or immoderate pump speed may cause ventricular collapse due to excessive suction. Suction causes decreased flow and may result in patient discomfort. Therefore, detection of this critical condition and immediate adaptive control of the device is desired. The purpose of this study is to evaluate and validate system parameters suitable for the reliable detection of suction. In vitro studies have been performed with a mock loop allowing pulsatile and nonpulsatile flow. Evidence of suction is clearly shown by the flow waveform reported by the implanted flow probe of the system. For redundancy to the implanted flow probe, it would be desirable to use the electronic motor signals of the pump for suction detection. The continuously accessible signals are motor current consumption and rotor/impeller speed. The influence of suction on these parameters has been investigated over a wide range of hydrodynamic conditions, and the significance of the respective signals individually or in combination has been explored. The reference signal for this analysis was the flow waveform of the ultrasonic probe. To achieve high reliability under both pulsatile and nonpulsatile conditions, it was determined that motor speed and current should be used concurrently for suction detection. Using the amplified differentiated current and speed signals, a suction-detection algorithm has been optimized, taking into account two different working points, defined by the value of the current input. The safety of this algorithm has been proven in vitro under pulsatile and nonpulsatile conditions over the full spectrum of possible speed and differential pressure variations. The algorithm described herein may be best utilized to provide redundancy to the existing flow based algorithm.

An automatic detection system for ventricular collapse was developed and tested in a first clinical trial as part of a physiological speed control concept for axial flow pumps. From this clinical experience, and based on the acquired data during this trial, an optimization of the developed system was performed. An already-existing database of 784 individual cases was extended. For harmonization of this database an additional 412 snap files were extracted from continuous data recordings and classified manually using a standardized procedure. The already-developed and clinically tested algorithms were supplemented by one additional indicator derived from a preexisting criterion. One threshold value was replaced by application of a numerically optimized nonlinear characteristic curve dependent on heart rate. Finally, in a multidimensional optimization process of the entire suction detection system, 7 individual indicators were adjusted by using 17 independent threshold values. The optimization criteria were applied using a three-level hierarchical system. Within the final database consisting of 1196 snap shots the overall amount of maldetections could be reduced to 23 cases including 5 false positive events (0.42%) and 18 false negative decisions (1.5%). By application of the clinical experience from the first clinical trial of a physiologic control system it became possible to optimize the sensitivity and specificity of the suction detection system to unprecedented accuracy.